Oral mucositis (OM) is a painful and debilitating consequence of cancer therapy, but its risk and outcomes during standard dose, multi-cycle chemotherapy are poorly described. The lack of such information compromises clinical decision making and design of clinical trials addressing prevention of OM. The objectives of this study are to 1) characterize the relationship between clinically significant OM during chemotherapy and its risk during subsequent chemotherapy cycles, 2) develop a predictive model of subsequent cycle risk and to assess the model's predictive value, and 3) to describe associations between OM and clinical and patient-reported outcomes. OM risk and outcomes will be studied in 177 patients with colorectal or breast cancers or non-Hodgkins lymphoma during 1121 cycles of chemotherapy. The chemotherapy regimens reflect the current standard of care for these cancers, worldwide (FolFOX, FolFIri, TAC, AC+T, and CHOP). Data will be obtained from an existing database of OM risk and outcomes created during a multi-center, prospective, longitudinal study of cancer treatment-induced OM. These objectives were not posed during the original study. However, the data are ideally suited to answering this question because patient reported OM and outcomes were collected daily during multiple cycles of chemotherapy using previously validated tools. Risk of clinically significant OM (percentage of patients or cycles affected) will be estimated, per cycle, then modeled using multivariate techniques. Subsequent cycle risk will be estimated, conditioned on previous cycle risk and severity, and modeled in similar fashion. Patterns of outcomes and resource utilization will be examined over multiple chemotherapy cycles. The clinical outcomes are weight loss, requirement for opioids, delays in planned therapy, dose reductions, and episodes of febrile neutropenia or bacteremia. The patient- reported outcomes are interference with activities of daily living, pain medication usage, and fatigue and quality of life scores. The healthcare resources are unplanned outpatient office visits, emergency room visits, and hospitalizations. Information resulting from this study will be instrumental in identifying solid tumor patients at risk of clinically significant OM or its serious outcomes for participation in clinical trials of new preventive agents. PUBLIC HEALTH RELEVANCE: The goal of this study is to determine whether the risk or severity of chemotherapy-induced injury to the mouth and throat tissues is higher among patients who have experienced this injury during a previous cycle of chemotherapy. This information is critical to clinical decisions about use of high cost preventive agents and frequency of clinic visits for side effect monitoring;it is also critical to design of clinical trials of new therapeutics.